The Unknown Ideal

The darned old S&P 500 valuation worksheet is showing US stocks pretty much fairly valued in relation to historic price-to-earnings ratios, assuming earnings hold up and assuming that historic mean P/E is not an idiotic benchmark for stock value to begin with.

Using another old, largely discredited model of the reasonableness of US stock prices, the Fed Model, which compares the "earnings yield" (i.e. earnings-to-price ratio) of the S&P 500 (5.89%) with the yield on the 10-year Treasury note (5.12%), stocks look a bit undervalued. Of course one of the reasons the Fed Model has been dismissed in recent years was the distortions to the 10-year's yield resulting from a global liquidity glut.

Taking a tack similar to the Fed Model, and comparing the S&P's earnings yield (5.89%) with today's aggregate yield of 10-year AAA-rated corporate bonds (5.92%), we get almost an even match in yields.

Obviously, with the rise in their yields, bonds seem a much better bargain today than they were with the 10-year Treasury at 4%. Are they as good a bet as US stocks? The uncertainty strengthens the case for diversification between stocks and bonds, and that case is much stronger than it has been in the recent past.

Posted by The Unknown at 3:12 PM | Permalink

First, money manager Seymour Lotsoff argues (though the idea is not new with him) that the shear size of indexed assets presents a threat to the longstanding, well documented superiority of index (passive) investing over active stock picking. Of course Lotsoff is an active manager. And the blindfolded dart-throwing monkey has, no doubt, beaten him before. And the monkey will probably beat Lotsoff again.

Next, Andre Perold joins the Fundamental vs. Cap-weighted (read: Siegel/Arnott vs. Bogle/Malkiel) debate with his draft paper "Fundamentally Flawed Indexing." As Douglas Appell reports for Pensions & Investments:

Mr. Perold said the key point of his paper is that if nothing is known about fair value, then any stock, regardless of capitalization, is just as likely to be overvalued as undervalued. Consequently, holding stocks in proportion to their market capitalization doesn’t systematically result in performance drag, he said.

Quant managers who have seen Mr. Perold’s paper say it’s a strong argument. To make the case for fundamental indexing, you have to presume that “large-cap stocks are overvalued, and you don’t know that,” said Eric H. Sorensen, president and chief executive officer of Boston-based PanAgora Asset Management Inc.

On that note, I can't recall whether it was Benjamin Graham or Warren Buffett who said, "God almighty doesn't know the proper price-to-earnings ratio for a common stock." Maybe it was Burton Malkiel. Probably all three of them said it at one point or other. But that's not going to stop me from pointing out that, based on its historical mean P/E ratio, the S&P 500 still looks to be fairly valued at its current level, if just a little pricier than a couple months ago.

True, stock prices have risen, but so have trailing earnings and estimates of future earnings.

Posted by The Unknown at 2:27 PM | Permalink

Mr. Broken Record here, with his updated US Stock Valuation Worksheet: I know the yellers on CNBC and elsewhere keep calling for a correction, but the current US large-cap stock valuations still look quite reasonable by historical standards.

Today the S&P 500 closed at 1451.19. Assuming a fairly conservative ("as reported") estimate of aggregate earnings of $92.25 for the index, that gives us a Price-to-Earnings (P/E) ratio of 15.73, right in line with the historical average of 15.68 since 1935.

Say we get a little harsher in evaluating earnings, and go with an S&P "core" earnings estimate of $85.70 for 2007. We're still looking at at P/E ratio of 16.93 for the year, still quite reasonable, especially in a period of modest inflation and low interest rates.

Project the high quality of "core" earnings to trailing 12-months earnings measures, and the P/E creeps up a bit to 18.16. Sorry, but that's still not a bad P to be paying for all that E. Bob Shiller would surely disagree, given that he likes to calculate his P/E based on 10-year trailing "smoothed" earnings, but Ed Yardeni would point out that investors don't discount the past, they discount the future.

US large-cap stocks still look reasonably priced at this date. I'm squinting really hard, but I can't see a bubble there (though I can certainly still see bubbles in plenty of neighborhoods in bond world - sorry, is that mixing metaphors?). A correction may come, but it would represent, more than anything else, a buying opportunity.

Posted by The Unknown at 2:55 PM | Permalink

It's awfully sweet that Goldman Sachs and Merrill Lynch are introducing investment products that attempt to replicate Hedge Fund Returns without the 20%-plus fees of Hedge Funds. But one must wonder what the point is, given that most hedge funds don't beat the S&P 500.

The race is on to commercialize synthetic hedge funds. For the past few years, academics have been discussing the possibility of replicating hedge fund performance using composites of market benchmarks. But it is the most recent research, including a paper from Massachusetts Institute of Technology finance professor Andrew Lo and Jasmina Hasanhodzic, an MIT Ph.D. candidate -- first published in Institutional Investor's sister publication, Alpha ("Attack of the Clones," June 2006) -- that has finally spurred securities firms into action.

Goldman, Sachs & Co. and Merrill Lynch & Co. are the first two to put academic theories on hedge fund cloning to the test. Both have products that they contend can deliver hedge-fund-like returns without investments in hedge funds. Using factor models similar to those proposed by academia, their indexes analyze historical monthly hedge fund returns and use weighted baskets of market indexes to replicate them.

Sure, one hears the argument from Hedge Fund investors that the S&P 500 isn't the appropriate benchmark for their pet investments. And one wonders what such people think the purpose of their investments might be.

Malkiel and Saha:

...hedge funds are marketed as an “asset class” that provides generous returns during all stock market environments and thus serves as excellent diversification for an all-equity portfolio. The funds have attracted close to $1 trillion of investment capital.

We showed that the practice of voluntary reporting and the backfilling of only favorable past results can cause returns calculated from hedge fund databases to be biased upward. Moreover, the considerable attrition that characterizes the hedge fund industry results in substantial survivorship bias in the returns of indices composed of only currently existing funds.

Why pay 1% in annual expenses for a dubious clone of a dubious hedge fund when you can pay a fifth of that and usually get better returns?

Posted by The Unknown at 12:48 PM | Permalink

Sorry it has been so long, gentle reader, my tender friend. I'll save my excuses for my coming appointment at the gates of heck.

Meanwhile Tom Adams has been kind enough to spell out what this morning's CPI report means for Series I savings bond investors:

December's CPI of 201.8 was up 2.5% from a year ago and it was up .15% from November's level. However, it's still below September's 202.9, which is the number I bond investors watch. That's because the I bond inflation component is based on the level of the CPI in March and September.

Investors already seem worried about the possiblity. Since the government's fiscal year began in October, new investments in Savings Bonds have barely recovered from the all-time lows seen over the summer.

Posted by The Unknown at 12:49 PM | Permalink

Today's close for the S&P 500 at a six year high of 1393.22 calls loudly for a long overdue update to the Valuation Spreadsheet.

Let's look at price-to-earnings ratios for that index and note that the P/E ratio for 2006 earnings of 16.35 is a bit beyond the average P/E of 15.38 that the index has turned in since 1935. So large-cap stocks seem to be edging toward the pricy side. But if we look at earnings estimates for 2007, a P/E ratio of 15.34 is produced, so large-cap US stocks still seem decently priced.

This simple P/E-comparison method of stock valuation is just that: simple. More sophisticated schemes are used by the likes of the bullish Jeremy Siegel, who compares a variety of earnings metrics to a variety of bond yields and still finds US stocks undervalued. His perennial counterpart on the bearish side, Robert Shiller, uses a "10-year smoothed" trailing P/E and finds shocking overvaluation still baked into the US stock cake.

For the finance dilettante typing this, simpler is better. I think stocks are possibly a little overvalued at this point, after a period of undervaluation, during which it was very good to be on the buying side. And even with stocks a little overvalued, I am comfortable continuing to dollar-cost average into my array of stock index funds for the foreseeable future.

Posted by The Unknown at 3:09 PM | Permalink

CNN reports on the the bond market's joy that some fiscal sanity might return to Washington, what with the anticipated demise of one-party Republican rule:

NEW YORK (CNNMoney.com) -- Treasury bonds rallied Tuesdays as investors bet that a change in control in Congress after the midterm elections could lead to greater fiscal discipline in Washington.

The bond market typically reacts well to more fiscal discipline since a lower budget deficit means the government has to sell fewer bonds to finance the deficit, which in turn helps support bond prices.

Posted by The Unknown at 12:43 PM | Permalink